HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
: Portfolio d’une balade infrastructurelle en zone rurale française
No full textInternational audienceThe deployment of telecommunications networks is generally planned by the State in a top-down manner: local actors must comply with implementation requirements. Using a survey method borrowed from infrastructural tourism, this portfolio starts from the points furthest from digital networks, in this case the last frontiers of digitisation in northern Ardèche, France, to highlight the tensions and local material arrangements that enable the installation of new infrastructure. It seeks to reveal and raise awareness of the material realities at the other end of the line: failed construction in wetlands, use of diesel generators to compensate for the lack of electricity connections, deployment in the footsteps of other industries and infrastructure, are just a few examples that help to understand this materiality.Le déploiement des réseaux de télécommunication est généralement pensé par l’État dans une logique descendante : les acteur·ices locaux·les doivent composer avec un impératif de mise en œuvre. Grâce à une enquête empruntant au tourisme infrastructurel, ce portfolio part des points les plus éloignés des réseaux numériques, dans ce cas les derniers fronts de numérisation en Ardèche du Nord, pour mettre en évidence les tensions et arrangements matériels locaux permettant l’implantation des nouvelles infrastructures. Il cherche à donner à voir et à rendre sensible aux réalités matérielles à l’autre bout de la ligne : construction ratée sur zone humide, usage de génératrices au fioul pour pallier l’absence de raccordement électrique, déploiement dans les traces laissées par d’autres industries et infrastructures, ne sont que quelques exemples pour comprendre cette matérialité
Invisibilisation Vibratoire dans des Plaques Minces par Approximation en Composites
No full textMatériaux poreux et métamatériaux acoustiques; GAPSUS - Acoustique Physique, Sous-Marine et Ultra-Sonore: GVB - Vibro acoustique et Contrôle du Bruit: GABE - Acoustique du Bâtiment et de l'EnvironnementNational audienceLe cloaking permet de rendre une inclusion dans un milieu indétectable par une onde se propageant dans ce milieu. Un domaine aux propriétés singulières est défini autour de l'objet à camoufler, de sorte que l'onde le « contourne » sans interaction. Les premiers résultats en électromagnétisme montrent la possibilité d’un tel phénomène par l’utilisation d’une transformation géométrique utilisée pour obtenir les équations de cloaking. Pour l’équation des ondes, cette méthode fonctionne puisque l’équation de cloaking correspondante est une équation d’onde généralisée au cas d’un milieu de propagation hétérogène et anisotrope. On parle d’invariance de forme. Contrairement à l’équation des ondes, l’équation des plaques minces de Kirchhoff-Love (KL) ne possède pas cette propriété. L’équation de KL transformée n’est pas interprétable physiquement à travers la théorie des plaques. Des résultats récents montrent néanmoins la possibilité d’atteindre l’invariance de forme en utilisant des transformations conservant l’aire. L’équation de cloaking résultante est donc une équation de plaque anisotrope, interprétable physiquement à travers la théorie des stratifiés symétriques. L’objet de cette étude concerne donc la résolution du problème inverse de l’obtention d’un matériau composite réel à partir de l’équation de cloaking. Les résultats numériques présentés démontrent la pertinence de la méthode développée et la capacité du design à supprimer le scattering induit par la présence d’un trou dans une plaque pour une large bande fréquentielle
Robust Explanations: The Case of Prime Implicants
No full textInternational audienceThis paper studies the problem of quantifying the robustness level of explanations. In particular, we propose to quantify the robustness level of each given explanation as the maximum level of perturbation on the parameters of a classifier below which an explanation remains valid. We derive theoretical results on the computational complexity of determining the robustness level in the case of linear models, which underpin the design of a log-linear time algorithm. We then apply the proposed notion of robustness level to analyze the robustness of commonly used specific types of prime implicants, including the shortest and the most robust ones. The insights are then leveraged to construct guidelines on scenarios where each type of explanations may be more beneficial
Analyse des effets des singularités induites par le procédé AFP au sein d'éprouvettes pleines et trouées en compression par dialogue essai/calcul
No full textInternational audienceThe AFP process induces unavoidable singularities such as gaps and overlaps. When singularities areplaced within coupons made of continuous carbon fibres impregnated with a thermoset matrix, thefailure properties under compression are modified. Experimental tests results are difficult to analysebecause of coupling effects that are probably due to structural effects that are difficult to characterisedespite a rich multi-instrumentation. To better understand the test results, an approach based on adialogue between the test results and the numerical simulations is developed. The method allows tocompare several meshed geometries and ply material laws in order to define the level of modellingdetails required to simulate the effects of singularities.Le procédé AFP induit inévitablement des singularités de type gaps et overlaps. Lorsque cessingularités sont placées au sein d’éprouvettes en fibres de carbone continues imprégnées de matricethermodurcissable, les propriétés à rupture en compression sont modifiées. Les essais expérimentauxde compression sont complexes à interpréter en raison de phénomènes de couplage liés à des effetsde structures difficiles à caractériser malgré une instrumentation riche. Pour aider à l’exploitation desdonnées d’essais, une démarche de dialogue essai-calcul est mise en place afin de simulernumériquement les essais. La démarche permet de comparer plusieurs géométries maillées et lois decomportement du pli afin de définir le niveau de détail de modélisation nécessaire pour simuler leseffets des singularités
Optimal Almost Sure Rate of Convergence for the Wavelets Estimator in the Partially Linear Additive Models
No full textInternational audienceIn this article, we examine a class of partially linear additive models (PLAM) defined via a measurable mapping Ψ:Rq→R. More precisely, we consider Ψ(Yi):=Yi=Zi⊤β+∑l=1dml(Xl,i)+εi,i=1,…,n, where Zi=(Zi,1,…,Zi,p)⊤ and Xi=(X1,i,…,Xd,i)⊤ denote vectors of explanatory variables. The unknown parameter vector is β=(β1,…,βp)⊤, and m1,…,md are real-valued functions of a single variable whose forms are not specified. The error terms ε1,…,εn are identically distributed with mean zero and finite variance σε, and they fulfill the condition E(ε∣X,Z)=0 almost surely. These models are broadly applicable in finance, biology, and engineering, where capturing intricate nonlinear effects is essential. We propose an estimation method that leverages marginal integration in conjunction with linear wavelet-based techniques to obtain estimators for the unknown components m1,…,md. Under suitable regularity conditions, we establish strong uniform convergence of these estimators, demonstrating that they achieve practically relevant convergence rates. Our theoretical results indicate that these estimators converge uniformly at rates that are favorable for practical applications, underscoring the adaptability and scope of this partially linear additive model
Limit Theorems for Kernel Regression Estimator for Quasi-Associated Functional Censored Time Series Within Single Index Structure
No full textInternational audienceIn this paper, we develop kernel-based estimators for regression functions under a functional single-index model, applied to censored time series data. By capitalizing on the single-index structure, we reduce the dimensionality of the covariate-response relationship, thereby preserving the ability to capture intricate dependencies while maintaining a relatively parsimonious form. Specifically, our framework utilizes nonparametric kernel estimation within a quasi-association setting to characterize the underlying relationships. Under mild regularity conditions, we demonstrate that these estimators attain both strong uniform consistency and asymptotic normality. Through extensive simulation experiments, we confirm their robust finite-sample performance. Moreover, an empirical examination using intraday Nikkei stock index returns illustrates that the proposed method significantly outperforms traditional nonparametric regression approaches
Customized Gaussian process for representing polycrystalline texture
No full textInternational audienceA customized Gaussian Process Regression (GPR) model is developed to reconstruct Pole Density Functions in texture analysis. The model integrates spherical-periodic distance measures with conventional stationary kernels, adapting the GPR framework to capture localized texture features. A key contribution is the introduction of a log-linear data transformation, which enforces the non-negativity of both interpolated function values and stochastic intervals, ensuring physically meaningful reconstructions. To assess the model’s effectiveness, a systematic investigation examines the impact of distance measures, kernel selection, and hyperparameter optimization using synthetic texture datasets, provided as part of this work, with evaluations focusing on reconstruction accuracy, feature preservation, and uncertainty quantification in comparison to the conventional spherical harmonics approach. GPR with a log-linear transformation, geodesic distance, and a Matérn nu =5/2 kernel, shows promise in achieving higher accuracy than traditional spherical harmonics for reconstructing non-negative pole density functions, while additionally providing confidence intervals for uncertainty quantification. Microstructure Texture Pole figuresDigital material representationGaussian Process RegressionKrigin
Defying Death: A Multi-Omics Approach to Understanding Desiccation Tolerance and Senescence in Eragrostis nindensis
No full textInternational audienceEragrostis nindensis is a resurrection grass capable of surviving near-complete desiccation. We compared non-senescent leaf tissue (NST) and senescent leaf tissue (ST) to investigate the cellular and molecular basis of desiccation tolerance and senescence. NST recovered fully after drying, while ST failed to regain viability. Integrated transcriptomic (using RNA-Seq), lipidomic (using LC-MS), and ultrastructural (Transmission Electron Microscopical) analyses revealed that NST maintain RNA processing, protein folding, and translational activity during desiccation. Lipidomic data and ultrastructure showed preferential accumulation of polyunsaturated triacylglycerols (TAGs) and lipid droplets in NST, supporting membrane protection and energy buffering. In contrast, ST showed cellular collapse, reduced oleosin protein accumulation, and signatures of senescence. These findings highlight the importance of post-transcriptional and post-translational regulation, as well as lipid metabolism, in preserving cellular integrity during desiccation in this species
Croissance in-situ de nanoparticules d'oxyde de fer dans une matrice de polyisobutylène : vers des nanocomposites multifonctionnels
No full textInternational audienc
A finite volume scheme for the local sensing chemotaxis model
No full textInternational audienceIn this paper we design, analyze and simulate a finite volume scheme for a cross-diffusion system which models chemotaxis with local sensing. This system has the same Lyapunov function (or entropy) as the celebrated minimal Keller-Segel system, but unlike the latter, its solutions are known to existglobally in 2D. The long-time behavior of solutions is only partially understood which motivates numerical exploration with a reliable numericalmethod. We propose a linearly implicit, two-point flux finite volume approximation of the system. We show that the scheme preserves, at the discretelevel, the main features of the continuous system, namely mass conservation, non-negativity of solution, entropy dissipation, and duality estimates. These properties allow us to prove the well-posedness, unconditional stability and convergence of the scheme. We also show rigorously that the scheme possesses an asymptotic preserving (AP) property in the quasi-stationary limit. We complement our analysis with thorough numerical experiments investigating convergence and AP properties of the scheme as well as its reliability with respect to stability properties of steady solutions